[c-nsp] necessity of nowadays

Chuck Church chuckchurch at gmail.com
Wed Mar 23 08:48:19 EDT 2016

UDLD works well when you've got a L2 switch with distributed processing,
such as a 6500.  We've had cases where a Sup was failing, perhaps due to
overheating in a failed air conditioned closet.  It failed to the point
BPDUs were no longer being sent, but forwarding was still working.  I guess
the SP wasn't happy, but PFC still forwarding.  Loop results.  UDLD fixed
that issue.  We did that prior to spanning tree loop guard existing.  I
think loop guard can replace UDLD in some cases.


-----Original Message-----
From: cisco-nsp [mailto:cisco-nsp-bounces at puck.nether.net] On Behalf Of
Sebastian Beutel
Sent: Wednesday, March 23, 2016 7:21 AM
To: cisco-nsp at puck.nether.net
Subject: [c-nsp] necessity of nowadays

Hi List,

    i've been pondering about the real need for udld nowadays, each time it
bites me in a case of false positive. At least since we have gigabit SFPs it
became almost impossible to willfully provoke an unidirectinal link: The
physical port allready detects missing light and goes down.
    Moreover, the main use of udld (prevent unidirectional loops in an stp
topology) has also lost importance since link aggregation has replaced load
balancing via multiple or per vlan stp topologys.
    That's why i am asking myself whether udld is a residue that nowadays
causes more harm than it prevents and should therefore not be used anymore.
At least on gigabit and faster links and if there are no really dumb media
converters involved.

What do you think?


More information about the cisco-nsp mailing list